DNA topoisomerase I and II (top1 and top2) are primary targets for cancer chemotherapy. Topo-isomerase poisons act by stabilizing enzyme-linked DNA breaks which can be detected as protein-associated DNA breaks in drug-treated cells (cleavable complexes). However, topoisomerases exist both in normal and cancer cells, and the mechanism(s) of drug selectivity for cancer cells remain(s) poorly understood. The goal of this project is to elucidate the antitumor mechanisms of topoisomerase poisons, and their selectivity for cancer cells. Secondary targets which are different in cancer cells could be used for drug development in association with topoisomerase inhibitors or other DNA damaging agents. We have studied the human colon and ovarian cancer cell lines from the NCI Anticancer Drug Screen to identify the parameters that are best correlated with cytotoxicity and could be used in the clinic to predict and monitor the response to topoisomerase inhibitors. Top1 protein levels appear comparable among cell lines and thus are not predictive, while top1 cleavable complexes are better correlated with cytotoxicity. Cleavable complexes are, however, not always predictive . Using cells lines with comparable cleavable complexes, we demonstrated the importance of the G2 checkpoint and replicon repair, downstream from the cleavable complexes for cytotoxicity. We have shown that the presence of p53 provides a survival advantage against camptothecin cytotoxicity. This may contribute to the selectivity of camptothecins for cancer cells with p53 mutations. We and others have proposed that apoptosis is a key response of cancer cells to chemotherapeutic agents, especially in leukemia cells exposed to topoisomerase inhibitors. We have shown that HL60 cells rapidly upregulate and downregulate cyclin B/cdc2 kinase activity before undergoing apoptosis after DNA damage. We have also set up an in vitro cell-free system to study the biochemical pathways of apoptosis. It appears that human leukemia HL60 cells induce a protease cascade including both ICE/CED3 and serine proteases in response to DNA damage following camptothecin treatment and that the protease(s) activate(s) preexisting nuclear pronuclease(s). New approaches aimed at triggering or suppressing apoptosis may provide new therapeutic strategies and help reduce drug side effects.